基于数据匹配和孤立点检测技术的审计证据获取系统
发布时间：2019-05-30 04:28
【摘要】： 随着信息技术的发展,以查账为主要手段的审计职业遇到了来自信息技术的挑战。因此,利用信息技术开展审计工作成为必然。但是,目前计算机辅助审计技术多是利用审计人员的经验和计算机查询技术相结合,从而无法处理庞大的数据库系统以及发现更隐蔽的审计证据,为此,本文将数据匹配和孤立点检测技术应用于审计证据获取中。 本文首先分析了当前计算机辅助审计技术的现状以及存在的问题,并介绍了数据匹配技术和孤立点检测技术的研究现状。根据审计证据获取需求,给出了审计证据获取系统框架和数据处理流程,并详细阐述了系统中各个组成模块的原理,介绍了系统的两个关键技术。 接着,详细研究了审计证据获取系统的一个关键技术:数据匹配技术。针对不同数据源的中的数据可能存在重复数据,从而造成财务舞弊,将数据匹配技术引用到审计证据获取中来检测相似重复数据,给出了数据中不同类型的字段相似度检测方法,重点研究了中文字符字段的相似检测算法,并进行了算法验证。 然后,详细阐述了审计证据获取系统的另一个关键技术:孤立点检测技术。由于审计证据多是一些孤立点数据,因此将孤立点检测技术引入来检测被审计数据中的孤立点数据,从而发现隐蔽的审计线索。根据被审计数据的特点,提出了一种先通过基于改进粒子群算法设计的数据划分方法对被审计数据进行数据划分再对被审计数据分组进行孤立点检测的方法,并进行了算法验证。 最后,初步实现了基于数据匹配和孤立点检测技术的审计证据获取系统,并采用实际数据对系统进行了验证,取得了良好的测试效果。
[Abstract]:With the development of information technology, the audit profession with checking accounts as the main means has encountered the challenge from information technology. Therefore, it is inevitable to use information technology to carry out audit work. However, at present, most of the computer-aided audit technologies make use of the experience of auditors and computer query technology, so that they can not deal with the huge database system and find more hidden audit evidence. In this paper, data matching and isolation point detection technology are applied to audit evidence acquisition. In this paper, the present situation and existing problems of computer aided audit technology are analyzed, and the research status of data matching technology and isolated point detection technology is introduced. According to the requirements of audit evidence acquisition, the framework and data processing flow of audit evidence acquisition system are given, the principle of each module in the system is described in detail, and two key technologies of the system are introduced. Then, a key technology of audit evidence acquisition system, data matching technology, is studied in detail. For the data in different data sources may have duplicate data, resulting in financial fraud, the data matching technology is applied to the acquisition of audit evidence to detect similar duplicate data. The similarity detection methods of different types of fields in the data are given, and the similarity detection algorithm of Chinese character fields is studied and verified. Then, another key technology of audit evidence acquisition system, outlier detection technology, is described in detail. Because most of the audit evidence is some outlier data, the outlier detection technology is introduced to detect the outlier data in the audited data, so as to find hidden audit clues. According to the characteristics of the audited data, a method is proposed to divide the audited data by the data partition method based on the improved particle swarm optimization algorithm, and then to detect the outliers of the audited data packets. The algorithm is verified. Finally, the audit evidence acquisition system based on data matching and isolated point detection technology is preliminarily implemented, and the system is verified by the actual data, and good test results are obtained.
【学位授予单位】：南京航空航天大学
【学位级别】：硕士
【学位授予年份】：2009
【分类号】：F239.1
本文编号：2488522
[Abstract]:With the development of information technology, the audit profession with checking accounts as the main means has encountered the challenge from information technology. Therefore, it is inevitable to use information technology to carry out audit work. However, at present, most of the computer-aided audit technologies make use of the experience of auditors and computer query technology, so that they can not deal with the huge database system and find more hidden audit evidence. In this paper, data matching and isolation point detection technology are applied to audit evidence acquisition. In this paper, the present situation and existing problems of computer aided audit technology are analyzed, and the research status of data matching technology and isolated point detection technology is introduced. According to the requirements of audit evidence acquisition, the framework and data processing flow of audit evidence acquisition system are given, the principle of each module in the system is described in detail, and two key technologies of the system are introduced. Then, a key technology of audit evidence acquisition system, data matching technology, is studied in detail. For the data in different data sources may have duplicate data, resulting in financial fraud, the data matching technology is applied to the acquisition of audit evidence to detect similar duplicate data. The similarity detection methods of different types of fields in the data are given, and the similarity detection algorithm of Chinese character fields is studied and verified. Then, another key technology of audit evidence acquisition system, outlier detection technology, is described in detail. Because most of the audit evidence is some outlier data, the outlier detection technology is introduced to detect the outlier data in the audited data, so as to find hidden audit clues. According to the characteristics of the audited data, a method is proposed to divide the audited data by the data partition method based on the improved particle swarm optimization algorithm, and then to detect the outliers of the audited data packets. The algorithm is verified. Finally, the audit evidence acquisition system based on data matching and isolated point detection technology is preliminarily implemented, and the system is verified by the actual data, and good test results are obtained.
【学位授予单位】：南京航空航天大学
【学位级别】：硕士
【学位授予年份】：2009
【分类号】：F239.1
【参考文献】
相关期刊论文 前10条
1 廖一兰;王劲峰;马家奇;戚晓鹏;;基于BPM-BM算法的地名数据匹配[J];测绘通报;2008年06期
2 陈竹;;新制度环境下对审计证据的认识[J];财会月刊;2007年33期
3 王志军,乐嘉锦;一种中文相似重复记录的检测方法[J];东华大学学报(自然科学版);2005年02期
4 陈伟,丁秋林,谢强;交互式数据迁移系统及其相似检测效率优化[J];华南理工大学学报(自然科学版);2004年02期
5 谢强;张磊;周良;;基于改进粒子群优化算法的Ontology划分方法[J];华南理工大学学报(自然科学版);2007年09期
6 俞荣华;田增平;周傲英;;一种检测多语言文本相似重复记录的综合方法[J];计算机科学;2002年01期
7 孙云;李舟军;陈火旺;;孤立点检测算法及其在数据流挖掘中的可用性[J];计算机科学;2007年10期
8 陈伟;张金城;QIU Ro-Bin;;计算机辅助审计技术(CAATs)研究综述[J];计算机科学;2007年10期
9 陈伟;Robin Qiu;刘思峰;;一种基于数据匹配技术的审计证据获取方法[J];计算机科学;2008年08期
10 李雪莹,刘宝旭,许榕生;字符串匹配技术研究[J];计算机工程;2004年22期
相关硕士学位论文 前1条
1 李俊娴;基于预处理技术的数据清理系统研究与实现[D];南京航空航天大学;2007年
，本文编号：2488522
本文链接：https://www.wllwen.com/guanlilunwen/shenjigli/2488522.html
